CN102782478B - For carrying out the method and apparatus of mist identification by spectroscopy - Google Patents
For carrying out the method and apparatus of mist identification by spectroscopy Download PDFInfo
- Publication number
- CN102782478B CN102782478B CN201180011836.6A CN201180011836A CN102782478B CN 102782478 B CN102782478 B CN 102782478B CN 201180011836 A CN201180011836 A CN 201180011836A CN 102782478 B CN102782478 B CN 102782478B
- Authority
- CN
- China
- Prior art keywords
- gasoloid
- image
- value
- mist
- headlight
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims description 59
- 238000004611 spectroscopical analysis Methods 0.000 title abstract description 5
- 239000003595 mist Substances 0.000 title description 75
- 238000001914 filtration Methods 0.000 claims abstract description 31
- 238000001228 spectrum Methods 0.000 claims description 35
- 238000005286 illumination Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 11
- 230000003595 spectral effect Effects 0.000 claims description 11
- 238000010521 absorption reaction Methods 0.000 claims description 10
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 230000003213 activating effect Effects 0.000 claims 1
- 238000012545 processing Methods 0.000 description 12
- 230000000694 effects Effects 0.000 description 11
- 230000008859 change Effects 0.000 description 10
- 230000000007 visual effect Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 7
- 230000004913 activation Effects 0.000 description 6
- 238000004040 coloring Methods 0.000 description 6
- 239000002245 particle Substances 0.000 description 6
- 238000012795 verification Methods 0.000 description 6
- 235000019504 cigarettes Nutrition 0.000 description 5
- 238000007689 inspection Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 239000000654 additive Substances 0.000 description 3
- 230000000996 additive effect Effects 0.000 description 3
- 239000000443 aerosol Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000002349 favourable effect Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 239000000779 smoke Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 239000000470 constituent Substances 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 125000001475 halogen functional group Chemical group 0.000 description 2
- 229910052724 xenon Inorganic materials 0.000 description 2
- FHNFHKCVQCLJFQ-UHFFFAOYSA-N xenon atom Chemical compound [Xe] FHNFHKCVQCLJFQ-UHFFFAOYSA-N 0.000 description 2
- 241000894006 Bacteria Species 0.000 description 1
- 235000002918 Fraxinus excelsior Nutrition 0.000 description 1
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 1
- 206010047571 Visual impairment Diseases 0.000 description 1
- 125000000218 acetic acid group Chemical group C(C)(=O)* 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 239000002956 ash Substances 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- -1 contrail Substances 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000003500 flue dust Substances 0.000 description 1
- 239000006260 foam Substances 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- ZZUFCTLCJUWOSV-UHFFFAOYSA-N furosemide Chemical compound C1=C(Cl)C(S(=O)(=O)N)=CC(C(O)=O)=C1NCC1=CC=CO1 ZZUFCTLCJUWOSV-UHFFFAOYSA-N 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 230000001795 light effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 230000005226 mechanical processes and functions Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000002105 nanoparticle Substances 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 231100000614 poison Toxicity 0.000 description 1
- 230000007096 poisonous effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 235000002639 sodium chloride Nutrition 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 239000002912 waste gas Substances 0.000 description 1
- 239000003643 water by type Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/47—Scattering, i.e. diffuse reflection
- G01N21/49—Scattering, i.e. diffuse reflection within a body or fluid
- G01N21/53—Scattering, i.e. diffuse reflection within a body or fluid within a flowing fluid, e.g. smoke
- G01N21/538—Scattering, i.e. diffuse reflection within a body or fluid within a flowing fluid, e.g. smoke for determining atmospheric attenuation and visibility
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1765—Method using an image detector and processing of image signal
- G01N2021/177—Detector of the video camera type
- G01N2021/1776—Colour camera
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/02—Mechanical
- G01N2201/021—Special mounting in general
- G01N2201/0216—Vehicle borne
Landscapes
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Lighting Device Outwards From Vehicle And Optical Signal (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention relates to and to identify by spectroscopy or density is determined or gasoloid of classifying.At this, image is taken by the video camera of vehicle, wherein, tried to achieve the first value of parameter in a first step by the first color filtering by least one image, in the second step not by or try to achieve the second value of same parameters by the second color filtering, described second color filtering is different from preceding color filtering, compares the value of trying to achieve at least two steps, and implements aerocolloidal identification according to the result of described comparison or density is determined or classifies.
Description
Technical field
The present invention is from the apparatus and method of classifying according to independent claims.
Background technology
In order to identify the dysopia especially caused by mist in vehicle, there are some in the prior art for identifying the method for mist.
Disclose a kind of device by EP1498721A1, it takes image by the video camera in vehicle along travel direction.The image of video camera described in following analyzing and processing: whether there is at least one lower than a brightness value, namely dark pixel.Scattered light effect when being existed by mist is drawn thus: all pixels have minimum brightness when there is mist and deducibility goes out to have mist thus.
Furthermore disclosed the camera chain for mist identification that some use in vehicle, these systems need active component (such as infrared LED), such as EP2020595A1.
Other method is the mist detection based on the identification of object and the identification of its distance where necessary.Edge strength in the edge strength of identified object and database compares and can derive visual field situation or rain or mist thus by EP1826648A2.This only just works when there is object and described object is described in a database.
EP2056093A1 describes a kind of more detailed method, observes that of image determines in region, grey value profile especially on local horizon in the process.But described method causes also may leading to errors during other weather conditions of same physical effect detection in the picture, possible example is the accumulated snow on landscape and street.
In addition there are a large amount of non-patent literatures with mist identification theme.
But the mist do not provided so far by formation method (video) detects, and do not need to determine distance (sighting distance), especially relative to the distance of other objects or at least one hypothesis about such distance.
Summary of the invention
In contrast to this, the method (device) with independent claims feature according to the present invention has the following advantages: can realize a kind of measuring method by the aerocolloidal spectrum signature of analyzing and processing, described measuring method when do not have illumination initiatively, the determination of distance or sighting distance or specific sensing mechanism only for gasoloid identification also no problem.Can passively be thrown light on by surrounding environment brightness or other light sources.
Based on the physical influence absorption that to be light improve when sweep gas colloidal sol, wherein light beam is absorbed at particle or molecule (such as water) place or is reflected.Such as, but described absorption is relevant to wavelength with disperse, recognizes: mist produces relatively cold light, i.e. high color temperature.Described effect and effect described in analyzing and processing subsequently can be identified by the color component observed in video image.
In addition advantageously, follow described optical relation can distinguish gasoloid and be arranged in other objects or the structure of image and additionally or alternatively also determine aerocolloidal type and/or density (concentration).At this, the image obtained by means of formation method or multiple image are implemented according to method of the present invention.Or there is according to the present invention the image that at least two have different colours information, or try to achieve described colouring information by an image, such as by known color extraction method by a RGB image or such as at the image obtained by R3l video camera by sheltering corresponding red pixel or non-red pixel (gray-scale pixels) tries to achieve described colouring information.Hardware technology measure (in the category of image taking) can be passed through and realize color filtering, respective color filters such as, before the parts being positioned at object lens or object lens or in the light path or by the see-through mask in the pixel of imager.In addition, the measure also can passing through software engineering (in the category of image procossing) realizes filtering, and its mode is, extracts desired Color Channel or color spectrum from coloured image.
The image of at least one process to be analyzed must take following scene: in described scene, gasoloid can be identified or visible, or rather, guarantee certain optical depth, dispersion effect or absorption effect are worked by its different spectral property in following degree: it works and is measurable at least one image described.Aim at (such as ground level/landscape) distant place and such as not aim at the cameras oriented on ground more meaningful, only have between video camera and ground and there is so little distance, make mist be only insignificantly visible.Described method is suitable for the application in motor vehicle, in described motor vehicle, at present usually video camera is provided with for other driver assistance functions, described video camera can additionally be born according to task of the present invention, and does not produce (important) additional firmware cost.
Should so filter at least one image now, make finally there are at least two (virtual) images, they comprise different colouring informations.According to aerocolloidal optical signature to be identified or spectrum signature, colour band to be filtered can be larger or less, also can be restricted to if desired in a single light frequency.Second (virtual) image under any circumstance must have the colouring information different from the first image.Correspondingly, filtrator also must differently construct.Alternatively, also can not use the second color filter and even can use the original image with all colours information (or half-tone information) where necessary.
If one or more aerocolloidal spectrum distinguishing characteristics needs detailed color spectrum, then described method not necessarily with described two images or only two color filterings apply, but can corresponding to the quantity multiplication of necessity and color gamut to be analyzed.Try to achieve a color parameter respectively by the image after described virtual image, i.e. color filtering, described color parameter is identical type but has following characteristics: have different values when a gasoloid or different gasoloid when extracting from corresponding image.If possible and be applicable to, then also can cancel the determination of virtual image when desired parameter such as can directly be tried to achieve when not having described step from original image.
To be compared to each other tried to achieve parameter value by suitable methods, thus can to identify whether there is gasoloid according to applicable cases, whether there is the gasoloid determined actually, there is which kind of gasoloid and/or described gasoloid has which kind of density (concentration).
When there is mist or the gasoloid with high diffusive or high reverse--bias, can advantageously avoid from dazzling the eyes, described from dazzle the eyes such as with far lighting thick fog and reflected light makes driver dazzling time produce.
Can advantageously determine gasoloid more accurately by method according to the present invention and avoid the explanation of error as often occurred so far, such as when snow cover track, described accumulated snow taken by video camera and image seems similar with visual field situation when having mist.
Advantageous extension and the improvement of the method illustrated in independent claims is can be implemented in by the measure illustrated in the dependent claims.
Advantageously, described method may be used for the gasoloid determined---and such as mist, its mode is, the red component in the first color filtering determination image and do not carry out the second color filtering, thus determines gray component at this.Red filter can be so constructed, and it is transparent for making it be only transparent or for the place at least partially of red spectrum whole frequency band for what a wavelength proper in red color range.The virtual gray level image with gray component directly can receive from original image according to device or when originally there is coloured image such as by the deletion of colouring information or on average determine.In the device of such as R3I video camera, this can be undertaken by sheltering of respective pixel, one of them mask filter red pixel and disappear other pixels and another mask filters gray-scale pixels and disappear red pixel, or there is not another mask, because element sensor directly provides gray-scale value.
The filtration possibility and analyzing and processing possibility do not mentioned can also be added on further here except described two color filterings.The blue filtered of adding can be considered, can the reasonableness test of result of implementation by described blue filtered.When having recognized mist, red component has experienced strong filtration, because mist absorbs and contrary correspondingly in blue spectrum less to absorb in described frequency range especially by force.
Advantageously, as the average brightness value of Selecting parameter image pixel.Light intensity (brightness) between the image can determining to carry out color filtering by described mode or other modes, to compare with its enforcement afterwards.Because gasoloid occurs planar diffusion phenomenon and cause the image of low contrast usually, so form the mean value of the brightness value of pixel.Try to achieve the spectral intensity distribution of the aerocolloidal image with supposition thus.
Alternatively, the average gradient of image pixel also can be selected as parameter.The gradient of two pixels is determined by the difference of its brightness value.Present average described gradient, to obtain the yardstick reflecting contrast thus.For mist, normally low gradient.High gradient is the mark that illuminated traffic mark board etc. is arranged in the area of image.Once there is high red component (due to its reflection characteristic) in described area, then the mist that the existence of object can lead to errors like this detects.The analyzing and processing of gradient standard stops thus and suppresses mist report when high gradient.
Known gradient of also carrying out in a plurality of directions in one or both directions, if desired between pixel is formed, not only can determine contrast in one direction, and can determine contrast in no direction.
Advantageously, need calculated performance relatively little compared with additive method for described method, described additive method such as must be implemented object and determine or necessary analyzing and processing image history.
Advantageously, at least two values by asking difference to carry out evaluating.Thus, the described comparison of two parameter values is implemented.Usually, described difference and a threshold value are compared, exceed or lower than during described threshold value identify or nonrecognition gasoloid.
Described difference (if desired when not having discrete threshold value to observe) also can be the tolerance for aerosol density.Density (concentration) is proportional with the droplet quantity of every volume unit.
The different threshold value of described difference or value scope can provide about aerosol type message and provide classification thus.Exactly use which threshold value or boundary or relation to depend on should by the corresponding aerocolloidal physical features determined for this reason.
Only when using described method night, such as, open distance light in order to prevent when there being mist, it is just enough to observe described difference.
Replace and ask difference also can consider to form business by two parameter values.The observation of business has the following advantages: also can carry out when different light situation (day/night) and the comparing of given threshold value, and compares that to achieve with brightness range irrelevant, because eliminate luminance level to a certain extent when asking business because described.Therefore, business describes the share of color in gray level image.
In addition, the variance of these values can also be formed.Described variance can be considered for the structurized of image and determine more fast.It is defined as the arithmetic mean of the squared-distance of large measurement (at this: pixel brightness value) and its mean value.Therefore, meaningfully, the variance variance of each desired Color Channel (or in RGB) of the variance of brightness value of the pixel in red image and the pixel of gray level image is calculated.When described value is greater than a threshold value, then can suppose a structurized scene, this does not relate to aerocolloidal mark.At this, especially consider the variance of gray-scale value and mean value, this is corresponding to the frequency in image.
In addition, also can implement to compare by means of the combination of grey value profile information.Relate to the distribution (when based on Gaussian distribution, then referring to variance and mean value) of gray-scale value or red value.
Enforcement relatively with above to differ from describe similar.
Advantageously, can be implemented according to inspection of the present invention with comparing of another threshold value by parameter value.At this, when the brightness of such as gray-scale value is greater than the brightness of red value, additionally inspection parameter value is greater than, namely exceedes the difference of the Second Threshold of first threshold relatively.This provides the information about the possible overexposure such as caused due to (self large light such as on the sign) undesirable reflection or active light source (headlight of the vehicle such as travelled in opposite directions) or large brightness, its when use only a threshold value be also categorized as gasoloid to possible errors.This such as occurs when pushing up vehicle below on the slope being positioned at front and come head-on and its headlight being suddenly visible.
But ask the modification of business by means of the brightness between gray-scale value and red value, this can not realize.
Advantageously, by means of one or more headlamp illumination gasoloid.These light sources have its oneself spectral property, and the physical influence that the spectral property thus by including based on it suitably selects the headlight be applicable to strengthen having described is to carry out gasoloid identification and to simplify identification thus.This or by selecting only applicable headlight to occur passively or by considering that when analyzing and processing the spectral property of headlight occurs on one's own initiative.
The spectral property of headlight shows the spectrum that headlight is launched, namely with what kind of wavelength of what kind of intensity.Can carry out consideration during analyzing and processing in the following way: such as matching threshold or select suitable comparative approach, or determine a parameter determined, it caters to the particular characteristics of selected headlight.
Advantageously, the spectrum of selected headlight and the convolution of aerocolloidal absorption characteristic to be identified is carried out.The spectrum of headlight is known, aerocolloidal spectrum to be identified (or contrary, absorption characteristic) be also known.Can in the hope of the spectrum of expection by convolution (decay of each wavelength approximate is multiplied---known from signal transacting), described spectrum occurs in the picture when there is corresponding gasoloid.For the discrete analyzing and processing such as illustrated, threshold value for each wavelength can be tried to achieve according to gasoloid and headlight now.Complete analysis of spectrum process is possible certainly equally.By considering that large modulation characteristic can be implemented to determine more accurately.If determine which kind of type the gasoloid in image is, then must may occur that the described convolution with the spectrum of headlight implemented by the gasoloid of (or to be identified) for each, obtain multiple threshold value and classification thus, and with in image actual occur comparing of spectrum can realize classifying.
Xenon headlight or the headlight be made up of LED element especially can be applicable to mist identification (being also applicable to other gasoloids if desired) well, because it has the blueness spectrum of reinforcement, described spectrum demonstrates obvious decay when the spectrum convolution with mist in red color range.This attenuation ratio is more obvious by the decay produced during white-light illuminating mist.
Meeting destination, only implementing asking for of parameter when considering a part for image.If successfully select suitable, significant image section, then reduce calculated load and computing time and resource can be realized to save.
When providing headlamp illumination as well as possible in described image section, the selection (area-of-interest---ROI) of described image section is favourable.When the large as far as possible area of the ROI that throws light on as far as possible by force, effect to be seen, difference between namely average red value and gray-scale value are maximum.Therefore terminator should be positioned at beyond ROI as much as possible.Terminator is by the predetermined line of headlight, and region transfers illuminated in described line is in dark, not illuminated region.
In an expanded configuration, image section can not only be selected in the picture regularly but also can mate ambient environmental conditions adaptively.It is contemplated that acetyl consider track model or topology (such as bend, top, slope or sink) alternatively and so select described part, make moving towards to implement gasoloid identification in significant region for further travelling.
Because it is significantly that described part is placed in illuminated region described above, so the tracking of described part is suitable for possible self-adaptation headlight (illumination distances regulating device, steering indicating light).
Image section also can move by relative direction dish/steering angle equably, and adaptive steering indicating light is like this too.At this, support following basic conception: the part illuminated the most by force by light be selected, because be the maximum efficiency of the spectrum suppression of red value (when there being mist) there.
Favourable by providing the soundness verification carrying out the gasoloid identification described according to the present invention about other data of gasoloid situation existed.Because can not debug identification or disturbing effect without doubt, other light sources that such as have unknown spectrum, that may act on to interference result, so such measure be significant.Aerocolloidal mistake/correctly identify can be limited further by such reasonableness test.
Can by known method, such as implement soundness verification by the method that illustrates in the prior art, or additive method, such as sighting distance is determined, wherein from determined sighting distance, gets rid of aerocolloidal existence.In addition, can determine to carry out soundness verification by gradient, wherein when image is (idealized, gray level image) in average gradient determine more than one threshold value time from nonrecognition gasoloid, this means certainly exist and determine the contrast rich (structure namely in image) that the gasoloid of density is inconsistent.In addition, in order to soundness verification, one or more external light source can be identified, and determine where necessary: whether these light sources have halation or halo.If not this situation, then gasoloid can not be there is between described light source and video camera.
In order to identify dazzle the eyes such as by mist cause from dazzling or such as caused by other vehicles other dazzle the eyes, such as can consider object identification, described object identification identifies other vehicles if desired.Reflection on the sign board thrown light on also may be mistakenly identified as mist.At this, such as, soundness verification by object identification is also helpful.
Advantageously, aerocolloidal identification meaningfully can be used in vehicle in order to multiple object.The aerocolloidal existence existed about gasoloid or determine or the information of its concentration can meaningfully for affecting security functions/comfortableness function/driver assistance function or drivetrain function on one's own initiative.
Hinder the gasoloid (such as mist) in the visual field if such as recognized, then can activate favourable Light distribation (when there being mist, such as vehicle front is dark, wide illumination).When not recognizing mist by means of method according to the present invention, although the whole accumulated snow (described accumulated snow can not distinguish with thick fog usually) on image display track, do not change Light distribation on the contrary or so regulate Light distribation, making Light distribation contribute to the wide visual field.In the case, can activate or keep high beam, certainly when considering other known rim conditions---as avoided the dazzling the eyes etc. of Reverse Commuting.
The automatic activation of mist taillight or mist headlight can be considered without any problems and stop using.
Example display is made as: how the activation for vehicle functions realizes soundness verification with light-operated.When recognizing mist, (difference the visual field) can abandon the change that headlight regulates, when by day instead of night time abandon the change that regulates.This such as can try to achieve by light sensor (also can to realize as function by described video camera or typically as independently sensor) and threshold value.In addition, when the speed of vehicle is lower and when being no more than threshold value, do not carry out the change of headlight adjustment.
If use is used for the method for photocontrol and simultaneously due to the less desirable effect (described effect may distort mist identification) that the appearance of traveling in opposite directions of vehicle is thrown light on, then this cuts little ice, because anyway will close distance light in said case.Therefore, these less desirable effects are resisted in photocontrol at least partly insusceptibly.
When from accident potentiality increase start (such as when the visual field is poor) time, can by suitable mode pre-tensioner or early warning security system---as belt pretensioner, air bag or brake assist/brake system.
When identifying the gasoloid determined---such as when cigarette, industrial gasses, comfort system can be transformed into circulating air or activation (air) filtrator from extraneous air.
The data staging that described driver assistance system in poor visual field situation is tried to achieve by video image can be more unessential or more disabled than the data of other sensors (such as radar, ultrasound wave) by driver assistance system.
Drivetrain can be affected as follows, make such as to improve rotating speed, stronger motor braking can be realized or produce louder running noise, the vehicle that described running noise can warn surrounding environment close.
Speed intervention and/or steering engagement can be implemented, to dodge the gasoloid of local or underspeed when there being mist or give a warning.These can in conjunction with the data from navigational system, and what described navigational system can evaluate the landform being positioned at front dodges possibility or passability.
Also can send aerocolloidal warning or urgent call, such as, when gasoloid is poisonous according to aerocolloidal type.
Concept explanation
Spectroscopy or spectral analysis are one group of observational techniques, and these observational techniques study electromagnetic radiation according to the spectrum (color decomposition) of light source (also can be image to be analyzed) and how material forms interaction.Research material when which frequency or wavelength can receive (absorption) or send the energy of (transmitting) photon or electromagnetic wave.Color decomposition can be implemented by one or more filtration.
Formation method required for the present invention can be realized: it must provide the image being suitable for color filtering by the camera chain with following characteristics.Described system can be typical RGB video camera or such as R3I video camera.In described R3I video camera, every 4 pixels are provided with a red mask, namely only receive red component (optionally pixel shader).Therefore, obtaining red image when only observing these pixels, obtaining gray level image when observing other pixels.Alternatively, also can use the camera chain with two monocular-cameras or multiple light path (having different filtration), or also can use a stereo camera, its 2 light paths are by different filtrators.
Gasoloid is the potpourri that a kind of suspended particle by solid-state or/and liquid state and gas are formed, and these suspended particles can be different sources.Elementary aerocolloidal particle is mostly from mechanical process or thermal process.In Secondary Aerosol, particle is formed by chemical reaction and/or by reactor product gathering on the nucleus of condensation by the material of gaseous state.Gasoloid can be natural and organic ingredients (pollen, spore, bacterium), natural inorganic composition (dust, meticulous dust, sand, cigarette, sea salt, water droplet), the products of combustion brought by people (cigarette of cigarette, cigarette, waste gas, industrial gasses, dense smoke, coal smoke or oil smoke (such as from automobile exhaust pipe), ashes or flue dust) or the nano particle manufactured by people.
Concept " hydrometeror (Hydrometeor) " uses in meteorology term, the upperseat concept of condensate water (word part " Hydro ") form of ownership, these condensate waters carry out observing (word part " Meteor ") in an atmosphere, all liquid state namely in water and freezing particle.What other of air were solid-state is called gasoloid with the constituent of liquid state, and when these constituents combine, water is inoperative or work at least hardly.The gathering of hydrometeror is moist steam, foam, contrail, mist, drizzle, rain, rain and snow mixed, cloud etc.
According to the present invention, the also claimed hydrometeror (or its assemble) when mentioning gasoloid.Whether hydrometeror distinguishes or aerocolloidal subset with gasoloid, is inconsistent in the literature sometimes.
In physics, vibrational energy is converted into other form of energy and is called decay.This such as shows the decay of signal, vibration or ripple.Electromagnetic wave is decayed in an atmosphere.When light absorption, the energy received not only is converted into heat, and can be machine-processed due to other---as fluorescence and by the scattering loss on gasoloid.At this, concept " absorption " and " decay " have similar implication.On the contrary, transmission corresponding to contrary, i.e. (depending on frequency) penetrability of light in gasoloid.
(part derives from online wikipedia)
Accompanying drawing explanation
Shown in the drawings and elaborate embodiments of the invention in the following description.
Accompanying drawing illustrates:
Fig. 1: the image taken by video camera with mist scene,
Fig. 2: for the block diagram of structure according to method of the present invention (device),
Fig. 3: the analysis of spectrum and do not have with mist scene,
Fig. 4: the process flow diagram carrying out photocontrol by means of mist identification according to the present invention.
Embodiment
Image 101 shown in Figure 1, as it can be taken in field of front vision by the video camera in vehicle.The mist scene with mist or thick fog 104 can be seen, lane markings 102 to extend in described scene and along with the increase of distance more and more invisible.Light cone 103 realizes the partial illumination of scene at du vehicule and realizes the illumination of mist 104 along with the increase of distance, and wherein light is returned by mist scattering and can mainly be occurred dazzling the eyes in dazzling range 105.A virtual line is defined, until the illumination of described virtual line is significant with border, the visual field 107.In practice, described line can not clearly imaging due to the continuous print optical characteristics of mist.According to the application examples of method of the present invention as the interested region of measured zone 106(can be limited to, ROI) image section on, described image section is positioned on the visual field or terminator idealizedly or comprises terminator.
Describe the structure according to method of the present invention (device) in fig. 2, wherein based on the video camera K of image 101 can be provided to implement to filter F1, described filtration provides (virtual) coloured image with desired colouring information.At this, F1 is the filtrator of the frequency component of preferred filtering not in red color range in described configuration.So-called red image is obtained by filtering F1, or red component.Alternatively, the second color filtering F2 is applied.But, illustrated for the embodiment of mist identification in be also suitable for original gray level image, thus must not carry out the second color filtering.The image of being tried to achieve by F1 or F2 by two can try to achieve the value of same parameters P respectively.Described parameter especially can be that the pixel of measured zone 106 is at its corresponding color filtering F1 and F2 or the average brightness value only in F1.In the case, a scalar value M is drawn respectively
g, M
r, it represents mean flow rate and for analyzing and processing in relatively V, such as, by two scalar M
g, M
rask poor D and inspection: described difference exceedes or lower than certain threshold value.When mist identification, the difference between the mean value of the brightness of gray level image and the mean value of the brightness of red image determine more than one value time, then supposition there is mist.
The above-mentioned inspection of the relative Second Threshold of difference is not shown at this, but can implements equally at this.
Suitable Threshold selection can be attempted by experience or physics is considered to draw.If detect mist now, then the opertaing device S be responsible for for photocontrol can correspondingly trigger headlight L.Especially, in the event of dazzling the eyes (high brightness namely in image or dazzling range 105) of supposition, then must check: whether exist mist and therefore existence dazzle the eyes and be not the scene (this is again desired) illuminated.When recognizing mist, such as can stop using high beam and/or activation fog lamp.
Show physics relationships graphically in figure 3.
Subgraph a) in the spectrum of xenon headlight has been shown, its intensity in blue spectrum, i.e. in the scope of higher frequency or more short wavelength λ or brightness I are higher than in low frequency (redness) scope.Distribute the mean value of the intensity in all frequencies can to described spectrum, this is corresponding to average gray value M
g.Redness according to the present invention is filtered and can so be constructed, and makes it only consider bands of a spectrum R, i.e. all frequency components of filtering not in bands of a spectrum R substantially.Also the brightness value distribution in described bands of a spectrum R can be given as the average brightness value M of the average strength in all frequencies in bands of a spectrum R
r.Two spectrum brightness value M
gand M
rdistinguished by difference D.Described difference changes, illustrating as following according to the existence of mist.Can be realized according to comparison of the present invention by difference described in analyzing and processing.Alternatively, other combinations that also can form business or describe in invention between these average brightness values.
At subgraph b) and d) in now illustrate the scene when there is mist continuity and at subgraph c) and e) in the continuity of described method when not having mist has been shown.
At subgraph b) in the spectrum transport property of mist has been shown, in low-frequency range, such as red color range, wherein there is higher decay (less transmissivity T) and lower decay occurred for higher frequency.At subgraph c) in the usual situation in the good visual field has been shown, namely do not decay.
If now by described absorption characteristic and subgraph a) in headlight compose convolution *, then obtain at subgraph d) and e) shown in intensity curve, wherein subgraph e) in curve corresponding to subgraph a) in primary curve because there is not any decay (1/T).The impact that caused by image subject or other light sources is not considered at this.These can come average due to the dynamic image change during forward travel.
At subgraph d) in illustrated carrying out spectrum change curve to be determined in the image of the video camera taken, when having as subgraph b in existence by corresponding to subgraph headlight a)) in the mist of attenuation characteristic time shooting image time.Tried to achieve the brightness M implemented at this by image corresponding to method according to the present invention
gand M
rthe determination of mean value.
Subgraph a) in (unaffected) average brightness that headlight is occurred drawn by the characteristic that headlight is known and be known.Can not be measured by video camera because it is in operation, so it can be stored/be configured for analyzing and processing and improve according to method of the present invention.
At subgraph d) with subgraph a) compared with mean value poor V amplification in can be clear that the effect of mist.
On the contrary, at subgraph f) in the Spectral structure felt in the picture has been shown, as unknown in its characteristic when headlight or not there is dominant spectral component or headlight does not connect (such as by day) time measured.At this, although the poor V of the mean value of brightness is measurable, can see in the application enhancing of the corresponding headlight time difference or increase, as subgraph a) and d) shown in.By having the enhancing effect of the headlight of known features, such as, can select the threshold value for mist identification larger, thus occur the mist identification of less mistake.Must consider on the contrary, these headlights can work, and make the spectral property also can with strengthening without mist scene with headlight taken in the picture, this described effect that may again decay.But empirical studies has confirmed this advantage.
Figure 4 illustrates a process flow diagram, as it can realize for photocontrol, be wherein used for mist identification according to method of the present invention.From starting point 401, described starting point such as can be transferred with the time interval determined repeatedly.Also element presented here and test can be selectively applied in, as the order inserting other significant points or change flow process on stream.
Whether implement sighting distance and determine 402, it such as can realize in the following way: such as check front sighting distance more than a threshold value, and hypothesis does not have mist 410 when exceeding.Such as can pass through its lane identification algorithm (track detection) and determine front sighting distance, when track (mark) is still visible in distance enough far away, described sighting distance is enough good.
If sighting distance is considered to too little, then implement contrast and determine 403, it such as can realize in the following way: implement gradient and determine, the average gradient wherein in gray level image must be over certain threshold value.In this case, then there is a corresponding contrast, namely consider there is not mist 410.Although if occur that (certainly) dazzles the eyes, then it can be caused by the sign board reflected.Therefore, the differentiation of the reason of dazzling the eyes can be realized.
On the contrary, if contrast is also too small, then can implement light source identification 404(external light source in addition) and check described light source whether to have halation or halo when there is such light source identification.If not this situation, namely light source is clearly and obviously visible, then draw there is not mist 410.
If contrary, then can continue the identification of applied spectroscopy mist according to method 405 of the present invention, described method can identify there is not mist 410 or there is mist 406 as described.Until described step 406 has operated know method for distinguishing for mist, be the method for photocontrol subsequently.
If recognized mist 406, then identify whether 407 inspections are nights just, and such as its mode is night, the threshold value that the light quantity of light sensor is determined more than.If daytime, then do not implement the change 411 that headlight regulates or regulates, because this may be expecting intentionally/regulating of driver when light activation.
If night, then implement rate test 408, described rate test represents safe gearshift approx, makes not carry out headlight when low speed (<50km/h) and regulates the change of 411, especially do not carry out the activation of high beam.If but met or exceeded a minimum speed determined, then should activate fog lamp 412 when there is mist.
If do not recognize mist 410, then such as can activate high beam when there is other rim conditions (such as there is no direction traffic).Can make recognizing mist or do not recognizing the Fast transforms Key dithering between mist, thus headlight adjustment can not change always or start flicker.Can correspondingly control mist taillight.
Claims (11)
1. for identify by means of at least one image (101) taken by video camera (K) of vehicle or density is determined or the method for (405) gasoloid (104) of classifying, wherein,
-in a first step, tried to achieve the first value of parameter (P) by the first color filtering (F1) by least one image described,
-in the second step, not by described first color filtering (F1) or second value of trying to achieve same parameters (P) by the second color filtering (F2), described second color filtering is different from described first color filtering (F1)
-compare the value that (V) try to achieve, and
-implement described aerocolloidal identification (401,406) according to the result (D) of described comparison or density is determined or classifies,
It is characterized in that, by means of headlamp illumination (103) described gasoloid (104), and consider the spectral property of described headlight (Fig. 3 a), implement the convolution (*) of the spectrum of described headlight and the absorption characteristic of at least one possible gasoloid (Fig. 3 b), and identify (406) described gasoloid (104) accordingly.
2. method according to claim 1, is characterized in that, described first color filtering (F1) determines red component (R, M
r), and do not determine gray component (M by the second color filtering (F2)
g).
3. the method according to any one of the preceding claims, is characterized in that, described parameter is average brightness value (M) or the average gradient of pixel.
4. method according to claim 1 and 2, is characterized in that, the comparison (V) of described parameter is by asking poor, asking business, asking variance or formed with the combination of grey value profile information.
5. method according to claim 4, is characterized in that, forms described comparison (V) by asking difference, and the correlativity forming result is compared by least two threshold values, wherein, when exceeding lower threshold value and lower than higher threshold value, identify or described gasoloid of classifying.
6. method according to claim 1, is characterized in that, the result of described comparison (D) experience is proceeded second compares, wherein, do not exist in aerocolloidal situation implement compare with second of the described expected results compared.
7. method according to claim 1 and 2, is characterized in that, only tries to achieve described parameter when considering a part (106) of described image.
8. method according to claim 7, it is characterized in that, the described part of selection like this (106), makes in a described part, realize illumination as well as possible by described headlight (103) and/or follow the tracks of a described part corresponding to steering angle and/or regulate a described part corresponding to the surface topology of the scene being arranged in described image (101).
9. method according to claim 1 and 2, is characterized in that, by the rationality of other data (402,403,404) checking gasoloid identification (405).
10. the application of the method according to any one of the preceding claims, for regulating the illumination distances of described headlight (L) or Light distribation (411,412) and/or activating or affect other security functions/comfortableness function/driver assistance function or drivetrain function.
11. for identify by means of the image (101) taken by video camera (K) of vehicle or density is determined or the device of (405) gasoloid (104) of classifying, wherein,
-try to achieve the first value by the first color filter (F1) by parameter determination unit (P) in a first step by described image,
-by identical parameter determination unit (P) in the second step not by or try to achieve the second value by the second color filter (F2), described second color filter is different from aforesaid color filter (F1),
-comparer (V) compares the value of trying to achieve at least two steps,
-implement described aerocolloidal identification according to the result (D) of described comparison or density is determined or classify (401,406),
It is characterized in that, by means of headlamp illumination (103) described gasoloid (104), and consider the spectral property of described headlight (Fig. 3 a), implement the convolution (*) of the spectrum of described headlight and the absorption characteristic of at least one possible gasoloid (Fig. 3 b), and identify (406) described gasoloid (104) accordingly.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102010002488.0 | 2010-03-02 | ||
| DE102010002488A DE102010002488A1 (en) | 2010-03-02 | 2010-03-02 | Method and device for detecting fog by means of spectroscopy |
| PCT/EP2011/051026 WO2011107308A1 (en) | 2010-03-02 | 2011-01-26 | Method and apparatus for detecting mist by means of spectroscopy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102782478A CN102782478A (en) | 2012-11-14 |
| CN102782478B true CN102782478B (en) | 2015-12-09 |
Family
ID=43902820
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180011836.6A Active CN102782478B (en) | 2010-03-02 | 2011-01-26 | For carrying out the method and apparatus of mist identification by spectroscopy |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US9297754B2 (en) |
| EP (1) | EP2542881B1 (en) |
| JP (1) | JP5575276B2 (en) |
| CN (1) | CN102782478B (en) |
| DE (1) | DE102010002488A1 (en) |
| WO (1) | WO2011107308A1 (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011017644A1 (en) * | 2011-04-28 | 2012-10-31 | Robert Bosch Gmbh | Method for controlling a headlight system of a vehicle and device for controlling a headlight system of a vehicle |
| DE102011017649B3 (en) | 2011-04-28 | 2012-10-11 | Robert Bosch Gmbh | Method and device for detecting an intensity of an aerosol in a field of view of a camera of a vehicle |
| JP5600332B2 (en) * | 2012-03-29 | 2014-10-01 | 富士重工業株式会社 | Driving assistance device |
| US9396528B2 (en) * | 2013-03-15 | 2016-07-19 | Digitalglobe, Inc. | Atmospheric compensation in satellite imagery |
| KR102089106B1 (en) * | 2013-06-19 | 2020-03-16 | 현대모비스 주식회사 | Method for Detecting Fog for Vehicle and Apparatus therefor |
| US9514373B2 (en) | 2013-08-28 | 2016-12-06 | Gentex Corporation | Imaging system and method for fog detection |
| US9734425B2 (en) | 2015-02-11 | 2017-08-15 | Qualcomm Incorporated | Environmental scene condition detection |
| US9846927B2 (en) | 2014-05-20 | 2017-12-19 | Qualcomm Incorporated | Systems and methods for haziness detection |
| US9911235B2 (en) | 2014-11-14 | 2018-03-06 | Qualcomm Incorporated | Spatial interaction in augmented reality |
| US10026163B2 (en) * | 2015-02-25 | 2018-07-17 | Cale Fallgatter | Hydrometeor identification methods and systems |
| JP6655983B2 (en) * | 2015-12-22 | 2020-03-04 | 京セラ株式会社 | Detection system, detection method, and vehicle |
| CN110036277B (en) * | 2016-03-10 | 2022-01-28 | 技术研究和发展基金有限公司 | Gas sensing device and gas sensing method |
| DE102016006326A1 (en) * | 2016-05-24 | 2017-11-30 | Audi Ag | Linear polarized light emission through a vehicle headlight for use in a camera-based driver assistance system |
| JP7113375B2 (en) | 2018-07-11 | 2022-08-05 | パナソニックＩｐマネジメント株式会社 | Display device, image processing device and control method |
| CN113077422B (en) * | 2021-03-22 | 2023-08-15 | 浙江大华技术股份有限公司 | Foggy image detection method, model training method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004076231A2 (en) * | 2003-02-21 | 2004-09-10 | Gentex Corporation | Automatic vehicle exterior light control systems |
| CN1751391A (en) * | 2003-02-21 | 2006-03-22 | 金泰克斯公司 | Automatic vehicle exterior light control system |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5149786A (en) * | 1974-10-25 | 1976-04-30 | Kogyo Gijutsuin | Earozoruno sokuteihoshiki |
| US6587573B1 (en) * | 2000-03-20 | 2003-07-01 | Gentex Corporation | System for controlling exterior vehicle lights |
| JP2002286842A (en) * | 2001-03-26 | 2002-10-03 | Mitsubishi Heavy Ind Ltd | Color laser radar equipment |
| EP1498721A1 (en) | 2003-07-15 | 2005-01-19 | ELMOS Semiconductor AG | Device for recognition of fog, especially for a vehicle |
| US7602937B2 (en) * | 2004-06-08 | 2009-10-13 | International Electronic Machines Corporation | Image-based visibility measurement |
| US7379177B1 (en) * | 2005-08-04 | 2008-05-27 | Applied Vision Company, Llc | System and method for performing hard glass inspection |
| JP5022609B2 (en) | 2006-02-27 | 2012-09-12 | 日立オートモティブシステムズ株式会社 | Imaging environment recognition device |
| JP4321591B2 (en) * | 2007-01-11 | 2009-08-26 | 株式会社デンソー | In-vehicle fog determination device |
| JP2009025050A (en) * | 2007-07-17 | 2009-02-05 | Sumitomo Electric Ind Ltd | Quality discrimination device of visual field, quality discrimination method of visual field and computer program |
| FR2919727B1 (en) | 2007-08-03 | 2010-07-30 | Valeo Vision | METHOD FOR DETECTION BY A VEHICLE OF A VISIBILITY DISPERSING PHENOMENON |
| FR2923016B1 (en) | 2007-10-31 | 2009-11-20 | Valeo Vision | METHOD FOR DETECTING A VISIBILITY DISRUPTION PHENOMENON FOR A VEHICLE. |
-
2010
- 2010-03-02 DE DE102010002488A patent/DE102010002488A1/en not_active Ceased
-
2011
- 2011-01-26 WO PCT/EP2011/051026 patent/WO2011107308A1/en active Application Filing
- 2011-01-26 JP JP2012555341A patent/JP5575276B2/en active Active
- 2011-01-26 CN CN201180011836.6A patent/CN102782478B/en active Active
- 2011-01-26 US US13/581,562 patent/US9297754B2/en active Active
- 2011-01-26 EP EP11702594.0A patent/EP2542881B1/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004076231A2 (en) * | 2003-02-21 | 2004-09-10 | Gentex Corporation | Automatic vehicle exterior light control systems |
| CN1751391A (en) * | 2003-02-21 | 2006-03-22 | 金泰克斯公司 | Automatic vehicle exterior light control system |
Non-Patent Citations (2)
| Title |
|---|
| Cloud Determination of All-Sky Images under Low-Visibility Conditions;HUO ET AL;《JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY》;20091001;第26卷;全文 * |
| Techniques for Identifying Dust Devils in Mars Pathfinder Images;STEPHEN M METZGER S ET AL;《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》;20000301;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102782478A (en) | 2012-11-14 |
| WO2011107308A1 (en) | 2011-09-09 |
| EP2542881A1 (en) | 2013-01-09 |
| US20130057687A1 (en) | 2013-03-07 |
| EP2542881B1 (en) | 2024-03-13 |
| DE102010002488A1 (en) | 2011-09-08 |
| JP2013521481A (en) | 2013-06-10 |
| US9297754B2 (en) | 2016-03-29 |
| JP5575276B2 (en) | 2014-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102782478B (en) | For carrying out the method and apparatus of mist identification by spectroscopy | |
| US10967793B2 (en) | Systems and methods for detecting obstructions in a camera field of view | |
| US8786697B2 (en) | Apparatus for determining the presence of fog using image obtained by vehicle-mounted imaging device | |
| US8077921B2 (en) | Apparatus for determining the presence of fog using image obtained by vehicle-mounted imaging device | |
| JP5846485B2 (en) | Adhering matter detection apparatus and adhering matter detection method | |
| US8477192B2 (en) | Optical detection system for motor-vehicles having multiple functions, including detection of the condition of the road surface | |
| CN103493099B (en) | The method and apparatus of the intensity of the aerosol in the visual field for identifying vehicle camera | |
| JP2014228300A (en) | Road surface state estimation device | |
| CN106846394A (en) | Snow cover travel path surface appearance is detected | |
| CN109703555B (en) | Method and device for detecting obscured objects in road traffic | |
| Sallis et al. | Air pollution and fog detection through vehicular sensors | |
| DE102015112103A1 (en) | Detection device for detecting fog for a motor vehicle | |
| CN110307995A (en) | A kind of test method for autonomous driving vehicle marginal test | |
| JP6555569B2 (en) | Image processing apparatus, mobile device control system, and image processing program | |
| US9171216B2 (en) | Method and device for detecting fog at night | |
| CN105844595A (en) | Method of constructing model for restoring headlight in nighttime traffic video based on atmosphere reflection-scattering principle | |
| CN102589515A (en) | Foggy-weather distance measurement method and device thereof as well as distance pre-warning method and device thereof | |
| CN116324478A (en) | Method for determining a sensor degradation state | |
| Imtiaz et al. | A Gas Imaging Sensor System of Automotive Exhaust Plumes for Remote Emission Sensing Application Basing on Schlieren Imaging An Approach to Visualize and Size Vehicle Exhaust Plumes | |
| Hautière et al. | Daytime visibility range monitoring through use of a roadside camera | |
| Cowan et al. | Investigation of ADAS/ADS Sensor and System Response to Rainfall Rate | |
| CN117549890A (en) | Early warning anti-collision method, device and equipment for vehicle and computer readable storage medium | |
| CN112622919A (en) | Drunk driving automatic monitoring system and detection method | |
| CN112782090A (en) | Drunk driving automatic monitoring system and detection method | |
| Sahana et al. | Survey on Fog Detection using in-vehicle Cameras for Evaluation of Road Condition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |